SUMMARY In almost all ecosystems, metals occur in mixtures, this is especially true at mining sites. Accurate assessment of metal exposure and effects to humans and ecosystems must account for mixture effects, which could be additive, less than additive, or more than additive. However, current regulatory and management practices usually address individual metals, because adequate tools are not yet available for predicting bioavailability and toxicity of mixtures. Our long-term objective is to address this shortcoming by improving the ability to measure, understand, and predict bioavailability and toxicity of metal mixtures arising from environmental contamination. We have proposed three specific projects in this work. First, we will solve a major challenge in analytical measurement of metal bioavailability by developing the gellyfish sampler to obtain free metal concentrations in water/sediment systems. Experiments will be performed both in the laboratory and in field deployments of the sampler at the Central City/Clear Creek Superfund site. Secondly, we will further develop a model of the toxicity of metal mixtures to individual species of aquatic organisms based on a mixed- metal, multi-site biotic ligand model (MMMS BLM). This will be accomplished with laboratory D. magna toxicity tests using both simple solutions and water samples collected from the site. The third objective is to integrate the procedures and information gained from the first two objectives with measurements of more complex effects of metal mixtures on aquatic communities exposed to mixtures in stream microcosms set up in the laboratory. We will also conduct a natural experiment in a metal-contaminated stream and examine responses before, during and after installation of a treatment system that will decrease concentrations of the metals. Because some metals like Fe can be toxic to aquatic invertebrates and fish in streams but can also make other metals less bioavailable (i.e., be protective), we expect to see a hysteresis in the recovery of the stream as Fe loading is decreased by the treatment system. Ultimately, in this project we expect to make improvements in assessment methods, which will provide a better means for protecting human and ecosystem health.